<![CDATA[This study aims to cluster regencies/cities in West Java Province based on public welfare indicators using the K-Means Clustering and Hierarchical Clustering methods. The data used includes health, economic, population density, and average length of schooling indicators in 2023. Cluster quality evaluation was performed using the silhouette score. The results show that K-Means Clustering with five clusters yields the highest silhouette score of 0.219. For comparison, Hierarchical Clustering with the Ward Linkage method and eight clusters was chosen, having a silhouette score of 0.202, which is the largest among other Hierarchical Clustering methods. The identification of each cluster's characteristics in K-Means reveals areas with multidimensional challenges (Cluster 1), industrial areas with unemployment issues (Cluster 2), areas with high stunting prevalence despite good access to basic facilities (Cluster 3), densely populated urban areas with good welfare but high unemployment (Cluster 4), and areas with very high health complaints and low welfare (Cluster 5). K-Means clusters (except Cluster 4) tend to have a low average length of schooling, below 12 years. Consistency in cluster patterns was found between K-Means and Ward Linkage, especially in advanced urban areas and areas with multidimensional welfare challenges in southern West Java. These findings are expected to serve as a reference for the government and policymakers in formulating more targeted and effective development strategies.]]>
Copyrights © 2025
